Transmissible spongiform encephalopathies (TSEs) comprise a group of rapidly progressing, neurodegenerative fatal diseases that affect both humans and animals. TSEs have clinical and neuropathological characteristics which include devastating dementia, pyramidal and extrapyramidal signs with myoclonus, multifocal spongiform changes, astrogliosis, amyloid plaques, neuronal loss, absence of inflammatory reaction and are usually characterized by a long incubation period.
In animals, a commonly known example of TSE disease recognized for over 200 years, is scrapie, which is found in sheep and goats (McGowan 1922). Other animal TSE diseases have also been described, such as transmissible mink encephalopathy (TME, Marsh 1976), chronic wasting disease of mule deer and elk (CWD, Williams 1980), bovine spongiform encephalopathy (BSE, commonly known as “mad-cow” disease (Wells 1987), and the more recently described feline spongiform encephalopathy of domestic cats, pumas, and cheetahs (Wyatt 1991).
In humans, TSEs have been traditionally classified into Creutzfeldt-Jakob disease (CJD), kuru, Gerstmann-Sträussler-Scheinker syndrome (GSS) and fatal familial insomnia (FFI). Among them, Kuru has been described only in the Fore linguistic group of New Guinea. For many years after its first recognition in 1957, kuru was the most common cause of death among women in the affected population, but its occurrence has declined because of the cessation of cannibalism that had facilitated disease transmission. As of today, only a few cases still occur due to the long incubation periods typical of this condition.
Although these rare neurodegenerative disorders occur in about 0.5 to one person per million worldwide (Brown 1987), TSEs attracted considerable public attention because of the unique biology and concerns about a onset of the epidemic of a newly recognized bovine spongiform encephalopathy (BSE) and its potential effects on human. There is mounting evidence that through dietary exposure to BSE infected tissues, it has poses a serious threat to public health and has resulted in an increased number of incidents of a newly recognized variant form of CJD (vCJD). Until now, there have been more than 100 cases of vCJD reported, a majority which are located in UK.
It is believed that prions are the pathogenic agent causing TSE. Many efforts have been directed towards identifying the etiological agent that causes TSEs. Early on, the transmissibility of TSE disease had been experimentally demonstrated in many cases, kuru and CJD from humans to chimpanzees (Gajdusek 1966, Gibbs 1968), transmissible scrapie from sheep to sheep (Cuillé 1936) and across species to goat (Pattison 1957). The most significant breakthrough was the successful transmission of scrapie to mice, by Richard Chandler in 1961 (Chandler 1961). Chandler's discovery greatly facilitated TSE research by providing an experimental model that was cheaper and easier to manipulate. Although all of the above modes of transmission were demonstrated experimentally, the cause of recent BSE in cattle and new variant CJD in human (vCJD) was considered a consequence of dietary exposure to the mix of scrapie sheep carcasses rendered for animal feed in the case of BSE (Brown 1997), and to beef from cattle affected with BSE in the case of vCJD (Bruce 1997).
It was suggested that TSE diseases might be caused by “slow viruses” or viroids (Gajdusek 1977). However, the extreme resistance of scrapie infectivity to radiation, nucleases, and other reagents damaging to genetic materials are inconsistent with the “virus” theory. Moreover, the infectious TSE agent could tolerate very high levels of heat and high concentrations of formaldehyde (Pattison 1965) while still able to replicate with the incubation period varying from a few months to over a year (Alper 1966).
All these “unusual” characteristics of the TSE infectious agent led Dr. Stanley Prusiner to propose the concept of “prions” in 1982 (Prusiner 1982). Prion (PrP), which stands for nucleic acid-free proteinaceous infectious particle, is a glycoprotein present in humans and animals. In humans, it is encoded by PRNP on chromosome 20 (Robakis 1986). The cellular form of this protein (PrPC) has two N-link glycosylation sites and a GPI anchor at the C-terminus. It has been most commonly found in neurons, and, to a much lower extent, it has also been found in other cells such as leucocytes, monocytes and platelets (Holada 2000). Furthermore, a soluble form of PrP that lacks the glycolipid anchor was detected in murine and human serum. The transmissible scrapie disease form of the prion protein (PrPSc) is a protease resistant isoform of its cellular precursor and is predominantly found in brain. At much lower level, it has also been found in tonsil, spleen, and lymph nodes in vCJD patients (Parizek 2001). The conversion from PrPC to PrPSc is believed to be accomplished through a conformational change within the protein. Although there is still ambiguity concerning the mechanism of the conversion, much experimental evidence indicates that in the presence of PrPSc, normal PrPC, acting as a substrate, undergoes a conformational structure change, and becomes PrPSc. This process of propagation involves replicating the conformation of PrPSc in PrPC and results in PrPSc aggregation and amyloid rod formation, hence causing cell death (Hope 1986, Horich 1997). As a result of Prusiner's concept of the “prion” as an infectious agent responsible for scrapie disease, and by extension, that of all TSE diseases gave rise to the notion of what are commonly referred to as Prion diseases to describe a class of pathologies believed to be linked to this protein.
Characteristics of PrPC and PrPSc 
The major property that differentiates PrPC and PrPSc is their distinct conformation. The structural change from PrPC to PrPSc is most supported by a crucial conformational change, involving a substantial increase in the amount of beta-sheet structure of the protein, with possibly a small decrease in the amount of alpha-helix, indicated by circular dichroism and infrared spectroscopy (Pan 1993, Caughey 1991). The solution structure of a fragment of the mouse PrPC has allowed a direct determination of secondary structure content of a portion of PrPC (121-231) by NMR (Riek 1996).
Protease resistance is another characteristic that distinguishes PrPSc from PrPC. In cultured cells and brain or in samples from many patients with GSS, PrPSc is smaller than its cellular precursor PrPC. Even though cellular prion and scrapie prion are two isoform of same PRNP genomic product, PrPC is completely degraded by Proteinase K treatment while PrPSc undergoes only limited digestion. The digestion yields a form of protein referred to as PrP 27-30 in which the N-terminus has been removed. PrP 27-30 has been postulated to be the PrPSc core required for PrPC hosted PrPSc replication. The protease treated prion molecule, PrP 27-30 or PrPres, is tightly linked to scrapie infectivity (Gabizon 1988), and provides additional evidence that PrPSc is an infectious protein.
An additional attribute, perhaps linked to the significant increase in β-sheet structure and concomitant protease-resistance, is the observed difference in solubility between PrPSc and PrPC. While PrPC is a soluble protein, the PrPSc isoform is highly insoluble. Furthermore, PrPC is found attached to the surface of neurons through a GPI tail anchored into membrane (Shyng 1994) while PrPRes is found in the cytoplasm of affected cells (Taraboulos 1990), most likely associated with late endosomal and lysosomal compartments (Arnold 1995), and PrPSc is also localized in amorphous aggregates in enriched fractions from infected brain (Meyer 1986). Interestingly, a disease-associated mutant PrP, the PrP159stop mutant was found exclusively in nucleus (Lorenz 2002).
There is mounting evidence indicating a tight linkage between scrapie infectivity and PrP 27-30. Even in the purest samples, the estimated ratio of PrP molecules to infectious units is ˜104 to 105 (Horwich 1997, Bolton 2001). At such low levels of infectivity, it is possible that other components, co-factors, or covalent modifications, are required for infectivity. The transgenic studies on the susceptibility of mice expressing chimeric human-mouse PrPC suggest the presence of at least one host factor other than PrPC, tentatively termed factor X, which might function as a molecular chaperone in the formation of PrPSc (Telling 1995).
Other Molecules Associated with Prion Pathogen
About 15 to 20 strains of scrapie have been identified based on their incubation period and lesion patterns in the inbred mice. After a serially inoculation passage in inbred mice homozygous for a single PRNP genotype, all the scrapie strains retained their original disease profile. These observations led investigators to question whether varied phenotypic strains were dominated by different conformation isoforms of same cellular prion precursor, a possibility suggested by conformation-dependent immunoassay (Safar 1998), or whether these strains were a result of various PrPSc associated molecules.
Many researchers have identified various nonprotein molecules that are bound to prion proteins. The precise biological and physiological roles remain the topic of further investigation. Copper and zinc have been demonstrated to bind to PrPC. In vitro, these divalent metals may contribute to prion superoxide dismutase (SOD)-like activity. Such SOD-like activity and copper content are dramatically reduced in scrapie-infected brain (Wong 2001).
In addition, prion rods, composed mainly of insoluble aggregates of the N-terminally truncated prion protein (PrP 27-30) are found to be associated with 1,4-linked glucose units. Sphingolipids, polysaccharide and other membrane components were also found in prion aggregates (Appel 1999, Klein 1998). The interaction between prion protein and lipid membranes could play a role in PrP conversion. For example, the negatively charged lipid membrane-inserted conformation of PrP is richer in β-sheet structure while the binding of PrP to raft-like membranes induces the formation of α-helical structure (Sanghera 2002).
In early 1990's, Snow et al, studying Gerstmann-Sträussler-Scheinker syndrome, Creutzfeldt-Jakob disease and scrapie, have documented the association of sulfated proteoglygan to the prion protein amyloid plaques (Snow 1990). In an immunohistochemistry study using heparan sulfate antibodies (anti-HS) and heparan sulfate proteoglycan antibodies (anti-HSPG), McBride has demonstrated the correlation and association between HSPG and abnormal PrP in scrapie-infected mice brain. This correlation and association was observed as early as 70 days post-infection and throughout the course of the disease (McBride 1998). In in vitro conversion from PrPC to PrPSc and in prion infectivity reconstitution experiments, sulfate glycans have been shown either to facilitate the conversion or to escalate infectivity (Wong 2001, Shaked 2001a). With recombinant GST::full-length prion and GST::prion fragment, Warner recently demonstrated direct binding of recombinant prion to heparin and heparan sulfate (Warner 2002). The peptide region 23-52 in prion sequence was positive in all HS and HSPG binding tests. Since the peptide failed to compete with full-length prion for binding to heparin, the author suggested that there might be another major GAG-binding site in intact PrPC. Another noteworthy observation is that GAGs from different species (bovine and porcine) or from different organs (lung, kidney and intestine) have shown different affinities for prion binding. The difference in affinity may be due to prion sequence itself, or may depend on the presence of particular sugar unit in the tested GAGs.
Through a mechanism that is perhaps different from that by which glycans participate in the conversion of PrPC to PrPSc, DNA could also convert cellular prion protein into β-sheet conformation (Cordeiro 2001). Nandi demonstrated that prion peptide 106-126 is the region that participated in the nucleic acid-prion complex association (Nandi 1998). Interestingly, not only was PK resistant amyloid aggregate obtained from the interaction between prion protein and nucleic acids, the nucleic acid morphology also changed to condensed globular structures, similar to nucleic acid structures induced by the HIV-1 NCp7 protein, but not to the structure induced by histones (Nandi 2001). Based on those in vitro conformation and conversion studies, it was hypothesized that DNA would act as a guardian of the PrPSc conformation as well as a catalyst to facilitate PrPSc conversion and aggregation (Cordeiro 2001).
Whether one accepts or rejects the “protein only” or “prion only” hypothesis, the effort to link inherited information to TSE disease or the search for genetic make up related to TSE disease has never stopped. The presence of a tightly bound RNA or DNA molecule in the prion particle was proposed to explain propagation of different strains of scrapie agent with distinct phenotypes in animals homozygous for the PRNP gene (Weissmann 1991). Analysis of highly purified scrapie prions by return refocusing gel electrophoresis revealed the small size of remaining nucleic acids, although the size of extracted nucleotides was too small to encode any meaningful protein (Kellings 1992). In a recent report, however, Narang indicated that animals inoculated with ssDNA purified from scrapie-hamster brains mixed with non-pathogenic prion developed clinical disease (Narang 2002). Based on his findings, he postulated that the “accessory protein” coded by the ssDNA may be involved in PrPC to PrPSc conversion. Although the role of nucleic acids in prion-associated disease is controversial, it is clear that PrPSc aggregates are tightly associated with these small molecules.
Infectivity and Transmissibility of Prion Diseases
Classic CJD in human has been grouped into three etiological types: sporadic (CJD), inherited (GSS or FFI), and acquired, which is very rare and includes diseases such as kuru and iatrogenic CJD. There is no hard evidence indicating any of CJD diseases is related to animal TSEs that may have crossed species barriers. The epidemic of kuru has provided the largest body of evidence of acquired human prion disease. Searching for risk factors and possible sources of infection in sporadic CJD patients revealed no significant correlation of disease to diet, blood transfusion or receiving other blood product. However, after intracerebral inoculation to mice, the infectivity in blood obtained from CJD patients indicated the possible presence of the CJD agent (Manuelidis 1985, Tateishi 1985).
BSE appears to have originated from dietary exposure. Nutritional supplements of processed meat and bone meal derived from scrapie disease infected carcasses were used to feed cattle livestock and other captive animals. In spite of BSE originating from scrapie, no case of de novo infection or cow-to-cow transmission has been reported.
There is mounting evidence, however, that links vCJD to BSE, The growing epidemiological data locates the majority of vCJD cases in UK where the overwhelming majority of BSE cases have also been reported. The link between vCJD and BSE is further supported by the neuropathologic evidence obtained from BSE-adapted macaques, the nearest model to humans (Bruce 1997), and from the study on inbred mice inoculated with the agent causing BSE and vCJD (Lasmézas 1996).
Although no vCJD patient has been documented as a victim of human-to-human transmission, the close link between BSE and vCJD attracted considerable attention. Concerns about human infection have been based on the observation that PrPSc is readily detectable in BSE and vCJD lymphoreticular tissues but not in classic CJD (Hill 1997), followed by the presumption that scrapie pathogen from sheep passage to cattle may have altered host range and become more adaptable to human. Experimental precedents for such behavior are well known: passage of mouse-adapted strains of scrapie through hamsters altered their transmissibility on back passage to mice (Kimberlin 1987, Kimberlin 1989); human strains of kuru or CJD did not transmit to ferrets or goats until passaged through primates or cats (Gibbs 1979); and a bovine strain of BSE did not transmit to hamsters until passaged through mice (Foster 1994). Alternatively, if BSE originated from a spontaneous mutation in cattle, experimental studies of species susceptibility to this new strain of transmissible spongiform encephalopathy (TSE) had not sufficiently advanced to predict that humans would not be susceptible.
In addition to CJD infectivity in blood described above, other TSE infectivity in blood has also been demonstrated in various experimental animals. Most blood for infectivity studies was obtained from TSE-adapted rodents such as mice and hamsters. The only exception was a study conducted in the sheep model. In this experiment, a sheep transfused with whole blood, taken from another sheep inoculated with BSE brain lysate, developed symptoms of BSE (Houston 2000, Hunter 2002). However, these experimental results yet need to be fully evaluated. The infectivity in blood has been established in rodent animals through intracerebral and intravenous transmission with mice-adapted BSE, mice-adapted vCJD and other rodent animal adapted TSE strains. Although the infectivity in lymphocyte-rich buffy-coat is greater than in plasma, it only accounts for relatively a small portion when compared to whole blood inoculums. The molecular definition of this infectious agent present in the blood is still under investigation. It is anticipated that finding of such infectious agent in blood would help us to better understand the relationship between PrPSc and TSE disease.
Study on human CJD and vCJD disease indicated that genomic susceptibility may yet be another factor that may influence the spread of TSE in humans. The majority of sporadic CJD patients were found to be homozygous for Met/Met or for Val/Val at codon 129 (Belay 1999). Nevertheless, all reported vCJD cases have been found to be homozygous for Met/Met.
The size and duration of vCJD epidemic still remains uncertain. Depending on the assumptions made and the modeling calculations employed, different predictions were proposed. One estimation of total vJCD predicts as few as 205 cases (Valleron 2001). On the other hand, another prediction for vCJD mortality for the next 80 years ranges from 50 to 50,000 if infection comes only from BSE. It could reach up to 150,000 if BSE is proven to infect sheep and if subsequently it is allowed to enter human food chain (Ferguson 2002). Although it is impossible to make accurate predictions if the necessary parameters are either mistaken or not available, one thing is certain that if vCJD infectivity is present in blood, any prediction will be an underestimate. In addition, vCJD has been proven to be a new disease entity and not simply the result of increased surveillance of CJD in humans (Hillier 2002).
Countermeasures have been taken by government to eliminate the spread of BSE incidence. Ruminant protein feed was banned in US and UK (1988). A series of measures have also been taken to prevent potentially infected meat from entering human food chain. To further reduce the human risk, FDA and CBER has issued a new policy in August 2001, which indefinitely defers any human blood donor who stayed cumulative ≧6 month during 1980-1996 in the United Kingdom (FDA 2001).
Diagnostic Assay for Prion Disease
Clinical symptoms of prion disease often overlap with those of other neuronal degenerative diseases that make diagnosis difficult. So far, PK resistant PrP 27-30 is the only protein marker linked to TSE disease. Therefore, the detection of this agent has become the focus of assay development. However the development of monoclonal antibody specific for PrPSc was extremely difficult, not only because pathogenic PrPSc isoform and normal cellular PrPC are two conformers of the same protein with an identical primary sequence, but also because the prion appears to be a weak immunogen. The only antibody reported to be able to recognize PrPSc specifically is not practically useful (Korth 1997). Other prion sequence-specific monoclonal and polyclonal antibodies are unable to distinguish PrPSc from PrPC. Nevertheless, these antibodies (such as 3F4, 6H4 described in U.S. Pat. No. 4,806,627 and EP0861900.) are still commonly in use for capture or for detection of prion protein in combination with sample treatment and separation techniques to isolate PrPSc from PrPC (Korth 1997, Kascsak 1987).
Since the outbreak of BSE in 1986, all commercially available tests for prion disease use, as their sample source, tissues taken from postmortem animals and humans. Among those, a tissue homogenate-based PrPSc assay, referred to as DELFIA (dissociation-enhanced lanthanide fluoroimmunoassay), was developed for the detection of scrapie prion (Barnard 2000 and a method described in US20020137114A1). It requires a protein denaturation step using GdnHCl, in combination with optional sample PK treatment and PrPSc enrichment by sodium phosphotungstic acid (NaPTA) precipitation. Since the transformation of PrPC to PrPSc is accompanied by the burial of epitopes near the N terminus of PrP, in DELFIA, monoclonal antibodies directed against the N-terminus of PrP are used to measure the difference of mAb binding affinity to the α-helical and β-sheet conformations before and after PrP denaturation (Peretz 1997). Another conformational-dependent immunoassay (CDI) combined with ELISA and fluorescence detection (Safar 1998, US 20010001061 A1, US20020001817A1) was described in conformation studies in PrPSc strains.
In a tissue distribution study of PrPSc in vCJD patients, an improved NaPTA precipitation was described to enrich PrPSc from brain and from other peripheral tissue homogenates (Wadsworth 2001). The modification employed endonuclease treatment to reduce sample viscosity prior to NaPTA precipitation. The recovery of PrPSc in the precipitated pellet was reported to be consistently greater than 90% while recovery of PrPC was about 5%. After PK digestion, the presence of PK resistant prion was verified in Western blot using 3F4 monoclonal antibody.
In another similar immunoblot assay, PK digestion was also used to eliminate PrPC. 6H4 was then used to determine the presence of PrPSc (Schaller 1999). Based on this first generation assay, a second-generation luminescence immunoassay was developed in which 6H4 was coated on plates as a capture antibody. The horseradish peroxidase (POD)-conjugated detection antibody used was a mouse monoclonal anti-PrP antibody, able to form a complex with PrP27-30 bound to 6H4 (Biffiger 2002).
The European Commission in 1999 evaluated 4 BSE test kits from different manufacturers (Moynagh 1999). They all used bovine brain tissue as a sample source, and all required a separate sample preparation procedure. Depending on the kit instructions, the brain tissue homogenate needed to be processed, including denaturation, PK digestion or PrPSc enrichment. The assay detection systems employed in DELFA, immunoblot, or in plate ELISA formats used either chemiluminescent or a colorimetric substrate.
In order to control the spread of the disease in the absence of a live-animal screening test, an extensive slaughter of cattle was carried out once an affected animal was identified within a herd. The urgency for a live animal diagnosis assay was reinforced when the first cases of variant Creutzfeldt-Jakob disease was reported in 1996.
Antemortem TSE diagnosis development presents three major difficulties: (1) insufficient sensitivity—Except in brain tissue, PrPSc concentrations in other tissues or fluids is considered to be very low. Therefore, a highly sensitive technique is required for detection. (2) Appropriate sample treatment—Any protein denaturation or PK digestion process may have a potential impact on pathogenic PrPSc structure, with the possibility of causing a false negative result. For example, it has been suggested that an intermediate form of PrPSc may not be PK resistant (Horiuchi 1999, Jackson 1999, Swietnicki 2000). And (3), the lack of PrPSc-specific antibodies and the incompletely characterized molecular relationship between the pathogenic agent and PrPSc in blood make it difficult to design an assay format for antemortem diagnosis.
A possible approach to boost the sensitivity is in-vitro amplification of PrPSc. It has been reported that when PrPSc was present, repetitive cycles of sonication could induce protease-sensitive cellular PrP to form protease resistant aggregates. The authors explained that in this “protein-misfolding cyclic amplification” (PMCA) process, sonication could disrupt newly formed aggregates and generate multiple smaller units for the continued formation of new PrPSc (Saborio 2001, WO0204954). At the end of 40 PMCA cycles, the sample was subjected to PK digestion and detected by immunoblot. It claimed that the amplification generated more than 30-fold protease resistant PrP. Since proteinase resistant PrP were generated at the expense of the normal prion protein as substrate through amplification cycles, a large quantity of same-species normal prion was required. It has not been demonstrated whether normal prion from another species could also work as substrate, or prion protein from a recombinant source or from sources other than brain tissue could be used. Such evidence would be useful when detection of vCJD is desired.
Immunohistochemistry of third eyelid lymphoid tissue has been described for preclinical diagnosis of ovine scrapie (O'Rourke 2000, U.S. Pat. No. 6,165,784, U.S. Pat. No. 6,261,790). Relying on a small surgical procedure, the assay makes use of sheep peripheral tissue, the third eyelid lymphoid for scrapie detection. The immunohistochemistry used a cocktail of pan-specific monoclonal antibodies to differentiate one isoform from the other. Following formalin fixation to reduce PrPC reactivity, the sample is subjected to formic acid and heat pretreatments which enhance the PrPSc reactivity. In spite of the fact that the assay is still tissue based and the observation that PrPSc displayed poor immunoreactivity in immunohistochemistry staining unless treated with denaturing agents, this antemortem preclinical diagnosis has made a step towards live-animal test as well as provided a way of identification of scrapie-affected sheep during the early, preclinical stage of scrapie.
In addition to the traditional identification of pathogenic prion by eliminating cellular prion followed by non-discriminatory anti-prion antibody recognition, other reagents were found to be able to differentiate PrPSc from PrPC, such as plasminogen and fibrinogen. The mechanism of interaction between these human blood component proteins and PrPSc is not clear. However, when immobilized on magnetic beads, plasminogen selectively precipitated PrPSc from brain homogenates of mouse, human, cattle and sheep. The evidence provided suggested that a property common to PrPSc of various species, rather than prion primary sequence or the specific tertiary structure of individual PrPSc molecules, could be responsible for binding to plasminogen (Fischer 2000, Maissen 2001). The application for the use of plasminogen and other serum/plasma proteins for the capture and detection of pathogenic prion protein has been described in WO0200713 and in US20010053533A1 (Aguzzi 2001).
Recent investigations have identified a new isoform of the prion protein in the urine of animals and humans with prion disease (Shaked 2001b, WO0233420A2). This isoform, referred to as UPrPSc by the investigators, was precipitable, PK resistant, and detectable only in infected individuals but not in normal controls. Most importantly, as indicated in their publication, UPrPSc appeared long before the clinical signs developed in inoculated hamsters. However, when UPrPSc isolated from scrapie hamster urine was inoculated back in normal hamster intracerebrally, it did not cause disease even after 270 days, well beyond the incubation period in which animal would develop clinical signs if comparable amount of brain derived PrPSc had been inoculated. It is not impossible that those hamsters, inoculated intracerebrally with UPrPSc, were still in a subclinical or carrier state. Moreover, PK-resistant PrP was not found in the kidneys, which implies that this UPrPSc could have originated from other organs and been transported to the urine via the blood. This important observation will undoubtedly lead to a better understanding of PrP metabolism.
Therefore there remains an unmet need for a better way to detect PrPSc and diagnose TSE in humans and animals. The aim of the present invention is to provide a non-intrusive way to isolate, concentrate and monitor the TSE disease-related pathogenic prion protein. The invention, including the use of selective anti-DNA antibody to bind the PrPSc through recognition of an associated binding partner, involves the discriminatory capture of PrPSc but not cellular prion protein. We provide evidence of a high affinity association of nucleic acid to PrPSc, and we demonstrate that such nucleic acids::PrPSc complex survived even after PK digestion and nuclease treatment.